Popular Science Monthly/Volume 7/August 1875/Telegraphic Determination of Longitude

From Wikisource
Jump to: navigation, search
Popular Science Monthly Volume 7 August 1875  (1875) 
Telegraphic Determination of Longitude
By F. M. Green


IN the construction of new charts for the use of navigators, as well as in the correction of old ones, the assignment of different latitudes and longitudes to the same point, by various authorities, has always been a source of difficulty and embarrassment.

The exact position of all prominent points on the coasts of the United States, as well as those of England, France, and other European nations, has been determined with great accuracy; but a large portion of the earth's surface is still very imperfectly and inaccurately laid down on marine charts.

The latitude of any point being determined directly by observation, and independently of the latitude of any other place, is less likely to be in error than the longitude, which can only be ascertained with reference to the meridian of some other place; being measured by time, is determined by the comparison of the local time with the time at some other place, the longitude of which is known.

Discrepancies in the results of observations for the determination of longitudes seem unavoidable with most of the methods usually employed, such as transportation of chronometers from place to place, observations of the relative positions of the moon and stars, and observations of occultations and eclipses.

Until the completion of telegraphic connection between this country and England, the exact longitude of the Washington Observatory was quite uncertain. A great many transfers of chronometers across the Atlantic had been effected by the Coast Survey at a great expenditure of labor and money. Yet the result of the latest expeditions differed from that deduced by Prof. Newcomb from moon-culminations by more than three and a half seconds of time, equal to nearly a mile, the final telegraphic determination lying between the two results.

In other parts of the world, however, the discrepancies are much greater. On the southern shore of the Caribbean Sea, an uncertainty of five or six miles exists with regard to many positions, and some of the islands in the Pacific Ocean have had longitudes assigned them by different surveyors within the last fifty years differing by as much as twenty-seven miles.

Where, however, chronometers have to be carried only a short distance from an established meridian, the results are much more accurate. In 1852, the longitude of Key West was measured by Coast Survey officers from Savannah (previously established by telegraph from Washington), and was found to be 81° 48' 30."7. In 1873, by telegraph, Washington to Key West 81° 48' 27."2. It will be seen that the difference between these results is only 3. "5, equal to about 100 yards, and that the statement lately published in Appletons' Journal, that the recent telegraphic determination showed the former position to be several miles in error, is incorrect.

Of late years the establishment of telegraphic connection between so many points of the earth's surface, both by submarine cable and by overland lines, has added to the modes of determining longitudes another, by far the most simple, elegant, and accurate.

This method can, however, only be used between places having telegraphic communication with each other; but the exact determination of these meridians renders easy the correction of errors in the longitude of neighboring places.

The establishment of differences of geographical longitude by the electric telegraph and of geographical latitudes by the zenith telescope constitute two of the most important improvements in practical astronomy of modern times, and both have had their origin in the United States. To the skillful and indefatigable astronomers of the Coast Survey and those of the corps of United States Engineers are due the introduction and perfection of the instruments and methods now employed, which make the results so accurate and the work so simple. "Among the very earliest of the astronomers to introduce this method of measurement was the lamented Captain J. M, Gilliss, U. S. N., who determined in this way the difference of longitude between Santiago and Valparaiso, Chili."

As soon as the Atlantic cable was laid, in 1866, the Superintendent of the Coast Survey took advantage of the opportunity to establish, by way of Newfoundland and Ireland, the difference between the meridians of the British Islands and those of the United States.

In 1869-'70, a similar determination was made by different observers through the French cable from Duxbury, Massachusetts, to Brest. Again, in 1872, the measurement was made through the same cable, using the island of St. Pierre, in the Gulf of St. Lawrence, as an intermediate station.

The exquisite accuracy of the results of these measurements is demonstrated by their accordance. Referring them to the station of the New York City Hall, the resulting longitudes are as follows:

1866— west of Greenwich 4h 56m 1s.71 equal to 74° 0’ 25".65
1870— " " 4h 56m 1s.70 " 74° 0' 25".50
1872— " " 4h 56m 1s.67 " 74° 0’ 25".05

The instruments in common use for making observations to ascertain the difference of longitude between two stations are, at each of the stations, a transit instrument, a break-circuit sidereal chronometer, and an electric chronograph; with the usual telegraphic sending and receiving instruments.

The transit instrument is a telescope, capable of being mounted accurately and firmly in an exact north-and-south line, so that the precise local time may be determined by the passage of well-known stars across the meridian.

The chronometer is adjusted to keep sidereal time and is furnished with an attachment by which the mechanism breaks an electric circuit every second.

Chronographs for the automatic registering of the exact time of any occurrence are constructed in various forms. Those generally used by astronomers in this country consist of a train of wheel-work driven by a weight, and causing a cylinder covered with a sheet of paper to make exactly one revolution in a minute.

A little carriage, to which a pen of peculiar construction is attached, moves upon wheels along the cylinder in the direction of its length, about one-tenth of an inch for each revolution of the cylinder, so that the pen records a perpetual spiral. The pen is so mounted as to have a slight lateral movement, and is so attached to an electro-magnet that, when the electric circuit in which it is placed is broken every second by the chronometer, which, with a small battery, is included in the same circuit, the mark made on the chronograph-paper, instead of being a straight line, will be broken at regular intervals as shown at a.

 a a a a b a a b a a

By means of a little instrument called, a break-circuit key, in the hands of the observer, and included in the same circuit, the electric current may be interrupted, causing the pen to make a similar mark as shown at b, on the occurrence of any event, such as the passage of a star across the wires of the telescope.

With a finely-divided scale the position of this arbitrary mark, with reference to the nearest second mark, may be accurately established, and the exact time accurately ascertained to within 120 of a second.

By means of these instruments, the error of the chronometer is found at each station with great accuracy, and, the times shown by the faces of the chronometers being compared by telegraph, the difference of time and corresponding difference of longitude are readily deduced.

The time occupied by an electric impulse to traverse the wire from one station to another, and act upon the telegraph-instruments, though generally very small, is too great to be neglected, but is easily ascertained and allowed for.

Suppose a to be a station, one degree of longitude east of another station b, and that at each station there is a clock exactly regulated to the time of its own place, in which case the clock at a will be, of course, four minutes faster than the clock at b. Let us also suppose that a signal takes a quarter of a second to pass over the telegraph-wire connecting the two stations.

Then if the observer at a sends a signal at exactly noon, by his clock,to 'b 12h 0m 0s
It will be received at b at 11h 56m 0s.25
Showing apparently a difference of time of 3m 59s.75
Then if the observer at b sends a signal at noon by his clock 12h 0m 0s
It will be received at a at 12h 4m 0s.25
Showing an apparent difference of time of 4m 0s.25

One-half the sum of these differences is 4 , which is exactly the difference of time and of longitude; and one-half of their difference is 0.25, which is exactly the time taken by the electric impulse to traverse the wire and telegraph instruments. This is technically called the "wave and armature time."

The error of each chronometer being ascertained by observations of stars at each station, and the difference of the chronometers being in this way shown by the exchange of signals, the difference of the local times, which is the difference of longitude of the two stations, is easily deduced.

Some English astronomers have objected that, where the line is, as is usual in long land-lines, divided into lengths connected by telegraphic repeaters, the time of transmission will not be the same in both directions, and that the same effect would be produced in a submarine cable having an imperfection or leak nearer one end than the other. Experiments, however, by the Coast Survey on the long line from Washington to San Francisco indicate that this objection, though theoretically true, is of no practical importance.

Upon land-lines the time-signals sent can be recorded directly on the chronograph by putting it in the telegraphic circuit; but, with submarine cables, the electric impulse transmitted is not strong enough to act upon the electro-magnets of the chronograph-pen.

For telegraphing with weak impulses over submarine lines a very beautiful device was invented by Sir William Thomson, and is now in general use.

To a delicately suspended magnet, surrrounded by a coil of fine covered wire, a small mirror is attached. From this mirror a beam of light from a lamp is reflected on a scale in a dark room. When no currents are being sent over the line, this beam remains at rest; but, when, at the sending station, either of two keys is pressed, a positive or negative current, as the case may be, is sent through the cable, and through the coil surrounding the magnet, causing the magnet with its mirror to turn and to deflect the ray of light to the right or left.

When the signal arrives and is perceived, the observer touches his chronograph-key, thus recording the time of its arrival.

The completion of the West India and Panama Telegraph Company's cable in 1873, and the certainty that serious errors existed in the geographical positions of many places in the West Indies and South America, caused Commodore R. H. Wyman, U. S. N., Hydrographer to the Navy Department, to turn his attention to the outfit of an expedition which should seek to determine with all possible accuracy the latitudes and longitudes of points connected by telegraph in that part of the world.

The authority of the Navy Department was readily obtained, and the necessary preparations were commenced in the spring of 1873.

In order that the work might be accomplished with economy, as small a vessel as possible was desirable, the Fortune, a strong iron tugboat of 300 tons, being selected and prepared. Although this little vessel carried the officers and men of the expedition safely, she was found to be too small to encounter heavy weather at sea with any degree of comfort.

The astronomical outfit was superintended by Mr. J. A. Rogers, of the Hydrographic Office, and was in all respects satisfactory.

The telescopes used were constructed at the repair-shop of the Hydrographic Office for the purpose, and were a combination of the transit instrument with the zenith telescope, a modification working admirably in practice, and first suggested by Prof. C. S. Lyman, of Yale College. These instruments were so constructed that the eyepiece was at one end of the horizontal axis, a prism at the junction of the axis and telescope-tube reflecting at a right angle the rays from the object-glass, thereby enabling the observer to direct the instrument upon stars of any elevation above the horizon without change of position.

The command of the expedition was given to Lieutenant-Commander F. M. Green, U. S. N., and it was intended that the work should be commenced in the winter of 1873-'74; but the non-completion of the instruments and the probability of trouble resulting from the Cuban outrages interfered with this plan, and the Fortune was temporarily employed as a tender to the squadron at Key West.

Upon the dispersion of the assembled squadron in April, 1874, Lieutenant-Commander Green was directed to complete a survey of the Mexican Gulf coast, commenced by the United States steamship Wyoming. This work employed the time till the following August, when the Fortune returned to Washington, and was at once refitted for the prosecution of the original design.

Fortunately for the success of the work, the services of Mr. Miles Rock, formerly of the observatory at Cordova, were secured as principal astronomical assistant; and the Fortune sailed on the 24th of November, 1874, from Hampton Roads for Jamaica.

Upon arrival at Kingston, definite arrangements were made with the manager of the telegraph cables, the gratuitous use of which had been offered very promptly and courteously by the London board of directors.

As it had been decided to commence the work by measuring between Panama and Aspinwall, the Fortune sailed for the latter place on the 9th of December, arriving on the 12th.

Portable observatories had been constructed, to shelter the instruments and observers, and were immediately set up at Panama and Aspinwall upon obtaining permission from the local authorities.

Throughout the work the same general system was pursued, and was briefly as follows: As soon as practicable after the establishment of a party at each station, the work was commenced by observing stars on five clear nights, from 8 to 10 p. m., and from 11 p. m. to 1 a. m. for determining the errors of the chronometers, and during the hour from 10 till 11 p. m. exchanging time-signals between the two stations. This was effected as follows:

Telegraphic communication being established between the observatories, the senior observer sent a preparatory signal at ten seconds before the completion of a minute by tapping his key several times in quick succession; then exactly at the even minute, pressing his key again for about a quarter of a second, and repeating this signal at intervals of five seconds till the completion of the next even minute. The hour and minute when the first signal was sent were then telegraphed to the receiving station and repeated to insure correctness. The time of arrival of these signals was recorded by the chronograph at the receiving stations, and five similar sets were exchanged in each direction, making sixty-five comparisons each way.

After five nights of this work, zenith telescope observations of pairs of stars were made on four nights for latitude.

In this way, during the winter, the stations of Panama, Aspinwall, Kingston, Santiago de Cuba, and Havana, were occupied, the exact difference of time between each station and the next and the latitude of each being ascertained. It was intended to continue the work to Key West, thus connecting with a Coast Survey station, but the occurrence of yellow fever among the crew of the Fortune, and the breaking out of that disease at Key West, caused the postponement of this measurement till the next season.

By combining these ascertained differences of time, and applying the result to a determined position, the longitude of each place will be decided with a very small limit of error.

In addition to the above observations, the exact habitual error of observing, or relative personal equation of the two observers, must be ascertained and applied to the result.[1]

The method of star-signals, or the comparison of the times at which the same star passes over the meridians of two stations, is seldom used now, and therefore is not described in detail.

The authorities of each country visited extended the most gratifying courtesy and assistance to the officers of the expedition. Especially was this the case in the island of Cuba, where a Spanish naval officer was detailed to assist in the work.

On the 5th of April last, all work practicable during this season being finished, the Fortune left Havana, completing her cruise by arriving at the Washington Navy-Yard on the 12th of the same month.

The computation of the numerous observations made during the past winter is now being prepared, and, as soon as completed, the results will be published.

Some improvements and modifications, which the experience of the past year has suggested, will be made in the instruments and outfit, and the same officers in a larger and more commodious vessel will leave the United States during the coming autumn, to continue the measurements through the Virgin and Windward Islands to the coast of South America.

Although the naval surveyors of nearly all maritime nations (particularly the English) are constantly at work perfecting the knowledge of the earth's surface, it is believed that this is the first systematic naval expedition for establishing by this method secondary meridians to which other positions may be referred.

In connection with their preparations for observing the recent transit of Venus, German astronomers have made some telegraphic measurements of differences of time in the East Indies; but the vast and constantly increasing net-work of cables nearly surrounding the earth will afford work for years to come, and will, in a way hardly contemplated by the projectors, add in a very great degree to accurate geographical knowledge.

Rule Segment - Span - 40px.svg Rule Segment - Span - 40px.svg Rule Segment - Flare Left - 12px.svg Rule Segment - Span - 5px.svg Rule Segment - Circle - 6px.svg Rule Segment - Span - 5px.svg Rule Segment - Flare Right - 12px.svg Rule Segment - Span - 40px.svg Rule Segment - Span - 40px.svg
  1. See Popular Science Monthly, vol. vi., p. 385.